Wan 2.1 I2V 14B 480P on 32 GB of VRAM
Yes. Q8_0 fits on 32 GB with 9.6 GB to spare. The image-to-video Wan 2.1, trained for 832×480. It has extra image-conditioning layers, so its files are bigger than the text-to-video 14B.
Runs well
Yes, with room to spare.
Q8_0 (18.1 GB) needs about 22.4 GB in use, so 9.6 GB of the 32 GB stays free: room for a larger image, a LoRA or a ControlNet.
Best fileQ8_0
File size18.1 GB
VRAM needed~22.4 GB
System RAM48 GB+
Memory map · 32 GB GPU22.4 GB / 32 GB
016 GB32 GB
Weights Q8_0 · 18.1 GBWorking memory · 3.5 GBReserve · 0.8 GBFree · 9.6 GB
calculated How this is counted.
Got an RTX 40 or 50? Swap Q8_0 for FP8 (16.4 GB): same quality class, computed natively there.
01What to download
| Part | File | Folder | Size | |
|---|---|---|---|---|
| Model | wan2.1-i2v-14b-480p-Q8_0.gguf Q8_0 · city96/Wan2.1-I2V-14B-480P-gguf | models/unet | 18.1 GB | Download → |
| Text encoder | umt5_xxl_fp8_e4m3fn_scaled.safetensors UMT5-XXL FP8 (scaled) · Comfy-Org/Wan_2.1_ComfyUI_repackaged | models/text_encoders | 6.7 GB | Download → |
| VAE | wan_2.1_vae.safetensors Wan 2.1 VAE · Comfy-Org/Wan_2.1_ComfyUI_repackaged | models/vae | 0.3 GB | Download → |
| Also needed | clip_vision_h.safetensors CLIP Vision H · Comfy-Org/Wan_2.1_ComfyUI_repackaged | models/clip_vision | 1.3 GB | Download → |
| Total download · keep about the same free on disk | 26.4 GB | |||
Alternatives: UMT5-XXL FP16 (11.4 GB). Sizes from Hugging Face, 2026-10-01.
System RAM: 48 GB or more. These files take about 26.4 GB of RAM while loaded, 32.4 GB with Windows and ComfyUI. What happens with less. calculated
02Every file, on 32 GB
| File | Size | Needed | On this card | Quality | Download |
|---|---|---|---|---|---|
| 16-bit SAFETENSORS · Comfy-Org | 32.8 GB | 37.1 GB | 5.1 GB too big | the original weights | Hugging Face → |
| Q8_0 ← GGUF · city96 | 18.1 GB | 22.4 GB | fits · 9.6 GB spare | practically identical to the original | Hugging Face → |
| FP8 SAFETENSORS · Comfy-Org | 16.4 GB | 20.7 GB | fits · 11.3 GB spare | practically identical to the original | Hugging Face → |
| Q6_K GGUF · city96 | 14.2 GB | 18.5 GB | fits · 13.5 GB spare | very close to the original | Hugging Face → |
| Q5_K_M GGUF · city96 | 12.7 GB | 17.0 GB | fits · 15.0 GB spare | close; small differences in fine detail | Hugging Face → |
| Q4_K_M GGUF · city96 | 11.3 GB | 15.6 GB | fits · 16.4 GB spare | good; some loss in fine detail and text | Hugging Face → |
| Q3_K_M GGUF · city96 | 8.6 GB | 12.9 GB | fits · 19.1 GB spare | noticeable loss of detail | Hugging Face → |
“Needed” adds 3.5 GB of working memory for Wan 2.1 I2V 480P and a 0.8 GB reserve. Sizes 2026-10-01.